The Cerebras Model Zoo is a comprehensive repository offering a range of deep learning models optimized for the Cerebras hardware. This collection is tailored to showcase the best practices for constructing models that exploit the robust capabilities of the Cerebras Wafer-Scale Cluster. With a focus on models developed in PyTorch, the Model Zoo provides detailed instructions for deploying neural network jobs on the Cerebras hardware, guiding users through compiling, validating, and training processes.
From version 2.2 onwards, the Model Zoo has been reorganized to enhance clarity and usability. If you’ve upgraded from any version prior to 2.2, see Directory Path Updates for more info.

​
Included Models

The Cerebras Model Zoo includes various models:
BERTBERT Classifier
BERT Named Entity RecognitionBERT Summarization
BloomBTLM
DiTDPO
DPRESM-2
ESM-2 ClassificationFalcon
GPT-2GPT-3
GPT-JGPT-J(Fine-tuning) Summarization
JAISLLaMA, LLaMA-2, LLaMA-3
Multimodal LLaVAMultimodal Simple
MistralMixtral
StarcoderRoBERTa
TransformerT5
To deploy your neural network jobs on the Cerebras Wafer-Scale Cluster, refer to the Quick start guide. This guide will walk you through the necessary steps to compile, validate, and train models from this Model Zoo using your preferred framework.

​
Model Portability

The Cerebras Model Zoo facilitates model portability, providing tools that aid users in adapting existing models or crafting new ones using Cerebras APIs. This includes a spectrum of user stages, from beginners to advanced users, with varying levels of integration and customization:

​
Beginners

Start with Cerebras data preprocessing tools and use model implementations found in the Cerebras Model Zoo.

​
Intermediate Users

Intermediate users can integrate their data preprocessing methods by referring to the section Create your own data preprocessing.

​
Advanced Users

Define your PyTorch model or code using the run function in the Cerebras Model Zoo and Supported Operations API.

​
Structure

The Model Zoo is designed to be user-friendly, offering an organized array of models, datasets, tools, and utilities. Key features include:
  • Reorganized models and datasets categorized by classes (NLP, Vision, Multimodal).
  • Registry APIs that enable querying of paths and supported combinations for each model.
  • Config classes designed to encapsulate parameters in YAML files, providing a clear class hierarchy and ensuring the validation of configurations used with models.
  • Enhanced data preprocessing tools for NLP models, significantly improving performance.
The structure of the Cerebras Model Zoo is illustrated in the diagram below:

Fig. 7 Cerebras Model Zoo structure

​
Modules

The principal modules in the Cerebras Model Zoo include:
ModulesDescription
modelzoo.configFor configuration classes and base-level configuration.
modelzoo.dataHosting data processing scripts and loaders for vision and NLP models
modelzoo.layersContaining layers and modules for model building and adaptation
modelzoo.lossesOffering a variety of loss functions for different model training phases
modelzoo.modelsImplementing a range of models across NLP, vision, and multimodal domains
modelzoo.data_preparationProviding tools for data preprocessing stages
modelzoo.toolsContaining utilities for model conversion and configuration
modelzoo.commonHosting common utilities across different models
modelzoo.trainerTrainer API that facilitates training and validating Model Zoo models
This redesign aims to streamline the user experience, making it easier for ML developers and researchers to explore, experiment, and develop solutions efficiently within the Cerebras ecosystem. The documentation further enriches this experience, offering examples and guidance on utilizing the Model Zoo’s Registry APIs and Config Classes, thereby enhancing the model development and deployment process.

​
Directory Path Updates

If you’re updating to the latest version of Model Zoo from any version prior to 2.2, the paths have been updated. See the table to compare the old paths with their corresponding new versions.
Old Path (relative to $MODELZOO)New Path (relative to $MODELZOO)
common/pytorch/*
common/run_utils/*
common/model_utils/*		common/
common/pytorch/layerslayers/
common/pytorch/input
common/input		data/
models/nlp/*
vision/pytorch/*
multimodal/pytorch/*		models/nlp/*
models/vision/*
models/multimodal/*
data_preparation/*data_preparation/
common/pytorch/model_utils/tools/
fc_mnist/fc_mnist/